Image Compression With Learned Lifting-Based DWT and Learned Tree-Based Entropy Models

TL;DR

This paper introduces a learned wavelet transform and entropy models for image compression, achieving superior performance over JPEG2000 and enabling efficient GPU implementation.

Contribution

It proposes a novel learned DWT via the lifting scheme and new entropy models that exploit dependencies, enhancing compression beyond traditional methods.

Findings

Outperforms JPEG2000 in compression quality.

Combining learned DWT with entropy models improves performance.

Achieves practical GPU encoding and decoding times.

Abstract

This paper explores learned image compression based on traditional and learned discrete wavelet transform (DWT) architectures and learned entropy models for coding DWT subband coefficients. A learned DWT is obtained through the lifting scheme with learned nonlinear predict and update filters. Several learned entropy models are proposed to exploit inter and intra-DWT subband coefficient dependencies, akin to traditional EZW, SPIHT, or EBCOT algorithms. Experimental results show that when the proposed learned entropy models are combined with traditional wavelet filters, such as the CDF 9/7 filters, compression performance that far exceeds that of JPEG2000 can be achieved. When the learned entropy models are combined with the learned DWT, compression performance increases further. The computations in the learned DWT and all entropy models, except one, can be simply parallelized, and the systems provide practical encoding and decoding times on GPUs.